March 7, 2014
David Reiley presents the results of a randomized experiment with 1.6 million customers measuring positive causal effects of online advertising for a major retailer. The advertising profitably increases purchases by 5%. 93% of the increase occurs in brick-and-mortar stores; 78% of the increase derives from consumers who never click the ads. This large sample reaches the statistical frontier for measuring economically relevant effects. Econometric efficiency was improved by supplementing experimental variation with non-experimental variation caused by consumer browsing behavior. This experiment provides a specification check for observational difference-in-differences and cross-sectional estimators; the latter exhibits a large negative bias three times the estimated experimental effect.